Input Reconstruction Reliability Estimation
|
|
- Rafe Greer
- 5 years ago
- Views:
Transcription
1 Input Reconstruction Reliability Estimation Dean A. Pomerleau School of Computer Science Carnegie Mellon University Pittsburgh, PA Abstract This paper describes a technique called Input Reconstruction Reliability Estimation (IRRE) for determining the response reliability of a restricted class of multi-layer perceptrons (MLPs). The technique uses a network's ability to accurately encode the input pattern in its internal representation as a measure of its reliability. The more accurately a network is able to reconstruct the input pattern from its internal representation, the more reliable the network is considered to be. IRRE is provides a good estimate of the reliability of MLPs trained for autonomous driving. Results are presented in which the reliability estimates provided by IRRE are used to select between networks trained for different driving situations. 1 Introduction In many real world domains it is important to know the reliability of a network's response since a single network cannot be expected to accurately handle all the possible inputs. Ideally, a network should not only provide a response to a given input pattern, but also an indication of the likelihood that its response is "correct". This reliability measure could be used to weight the outputs from multiple networks and to determine when a new network needs to be trained. This paper describes a technique for estimating a network's reliability called Input Reconstruction Reliability Estimation (IRRE). IRRE relies on the fact that the hidden representation developed by an artificial neural network can be considered to be a compressed representation of important input features. For example, when the network shown in Figure 1 is trained to produce the correct steering direction from images of the road ahead, the hidden units learn to encode the position and orientation of important features like the road edges and lane markers (See [pomerleau, 1991] for more details). Because there are many fewer hidden units than input units in the network, the hidden units cannot accurately represent all the details of an 279
2 280 Pomerleau SUp Ri&hI 30 Output Units 3Ox32 SelUlor Input Retina Figure 1: Original driving network architecture. arbitrary input pattern. Instead, the hidden units learn to devote their limited representational capabilities to encoding the position and orientation of consistent, frequently-occurring features from the training set. When presented with an atypical input, such as a road with a different number of lanes, the feature detectors developed by the hidden units will not be capable of accurately encode all the actual input features. Input Reconstruction Reliability Estimation exploits this limitation in representational capacity to estimate a network's reliability. In IRRE, the network's internal representation is used to reconstruct in the input pattern being presented. The more closely the reconstructed input matches the actual input, the more familiar the input and hence the more reliable the network's response. 2 Reconstructing the Input IRRE utilized an additional set of output units to perform input reconstruction called the encoder output array, as depicted in Figure 2. This second set of output units has the same dimensionality as the input retina. In the experiments described in this paper, the input layer and encoder output array have 30 rows and 32 columns. The desired activation for each of these additional output units is identical to the activation of the corresponding input unit. In essence, these additional output units turn the network into an autoencoder. The network is trained using backpropagation both to produce the correct steering response on the steering output units, and to reconstruct the input image as accurately as possible on the encoder output array. During the training process, the network is presented with several hundred images taken with a camera onboard our test vehicle as a person drives (See [pomerleau, 1991] for more details). Training typically requires approximately 3 minutes during which the person drives over a 1/4 to 1/2 mile stretch of road.
3 Input Reconstruction Reliability Estimation 281 Figure 2: Network architecture augmented to include an encoder output array. During testing on a new stretch of road, images are presented to the network and activation is propagated forward through the network to produce a steering response and a reconstructed input image. The reliability of the steering response is estimated by computing the correlation coefficient p(l, R) between the activation levels of units in the actual input image I and the reconstructed input image R using the following formula: p(i,r) = ir - I. II 01 OR where I and R are the mean activation value of the actual and the reconstructed images, IR is the mean of the set formed by the unit-wise product of the two images, and u/ and UR represent the standard deviations of the activation values of each image. The higher the correlation between the two images, the more reliable the network's response is estimated to be. The reason correlation is used to measure the degrees of match between the two images is that, unlike Euclidean distance, the correlation measure is invariant to differences in the mean and variance between the two images. This is important since the mean and variance of the input and the reconstructed images can sometimes vary, even when the input image depicts a familiar situation. 3 Results and Applications The degree of correlation between the actual and the reconstructed input images is an extremely good indicator of network response accuracy in the domain of autonomous driving, as shown in Figure 3. It shows a trained network's steering error and reconstruction error as the vehicle drives down a quarter mile stretch of road that starts out as a single lane path and eventually becomes a two-lane street. The solid line indicates the network's steering error, as measured by the difference in turn curvature between the network's steering response and a person's
4 282 Pomerleau CoJTelation Coeff"u:ienl = 0.92 Steering Error ii.pu;reoo~n&ror 0.04 ~ g -;; ~~ b.ov v... - v en... 'S e Intersection ~l L- ~ 000 O~ I... t ----One Lane Road Im.gel----~.. ~1 ~nll"' r--t.. Two Lane Road Images ~ 1.00 ~ c::~ o~ 'a ~b: jg~ ~ Figure 3: Reconstruction error obtained using autoencoder reconstruction versus network steering error over a stretch of one-lane and two-lane road. steering response at that point along the road. The dashed line represents the network's "reconstruction error", which is defined to be the degree of statical independence between the actual and reconstructed images, or 1 - p(i, R}. The two curves are nearly identical, having a correlation coefficient of This close match between the curves demonstrates that when the network is unable to accurately reconstruct the input image, it is also probably suggesting an incorrect steering direction. Visual inspection of the actual and reconstructed input images demonstrates that the degree of resemblance between them is a good indication of the actual input's familiarity, as shown in Figure 4. It depicts the input image, network response, and reconstructed input at the three points along the road,labeled A, Band C in Figure 3. When presented with the image at point A, which closely resembles patterns from training set, the network's reconstructed image closely resembles the actual input, as shown by the close correspondence between the images labeled "Input Acts" and "Reconstructed Input" in the left column of Figure 4. This close correspondence between the input and reconstructed images suggests that the network can reliably steer in this situation. It in fact it can steer accurately on this image, as demonstrated by the close match between the network's steering response labeled "Output Acts" and the desired steering response labeled ''Target Acts" in the upper left corner of Figure 4. When presented with a situation the network did not encounter during training, such as the fork image and the two-lane road image shown in the other two columns of Figure 4, the reconstructed image bears much less resemblance to the original input. This suggests that the network is confused. This confusion results in an incorrect steering response, illustrated in the discrepancy between the network's steering response and the target steering response for the two atypical images. The reliability prediction provided by IRRE has been used to improve the performance of the neural network based autonomous driving system in a number of ways. The simplest is
5 Input Reconstruction Reliability Estimation 283 Figure 4: The actual input. the reconstructed input and the point-wise absolute difference between them on a road image similar to those in the training set (labeled A), and on two atypical images (labeled B and C).
6 284 Pomerleau ' ~ L i\ I!'i, I!iJ a I t-' a :!I\ II I.. v, it,,':: \.:1 i!= \ f \ : f ~ "\." II II \ i. / \ ii I!i II I \...: 1 II ';. 1 'V: \\! y ~ OJ \.J, One Lane Net Rcconstuction Error TW~ line-net Rct;;n;tiiCtion Eir~; 0.20,., (\ \ :. \.-: :.,,. :. =, : ~. J...: '\i \_.: O.OOL.- I.. One Lane Images Two Lane Images -..j Figure 5: Reconstruction error of networks trained for one-lane road driving (solid line) and two-lane road driving (dashed line), to use IRRE to control vehicle speed. The more accurate the input reconstruction, the more confident the network, and hence the faster the system drives the vehicle. A second use the system makes of the reliability estimate provided by IRRE is to update the vehicle's position on a rough map of the area being driven over. When the map indicates there should be an intersection or other confusing situation up ahead, a subsequent sharp rise in reconstruction error is a good indication that the vehicle has actually reached that location, allowing the system to pinpoint the vehicle's position. Knowing the vehicle's location on a map is useful for integrating symbolic processing, such as planning, into the neural network driving system (for more details, see [pomerleau et al., 1991]). Figure 5 illustrates how IRRE can be used to integrate the outputs from multiple expert networks. The two lines in this graph illustrate the reconstruction error of two networks, one trained to steer on one-lane roads (solid line), and the other trained to steer on two-lane roads (dashed line). The reconstruction error for the one-lane network is low on the one-lane road images, and high on the two-lane road images. The opposite is true for the network trained for two-lane road driving. The reliability estimate provided by IRRE allows the system to determine which network is most reliable for driving in the current situation. By simulating multiple networks in parallel, and then selecting the one with the highest reliability, the system have been able to drive the Navlab test vehicle on one, two and four lane roads automatically at speeds of up to 30 miles per hour. 4 Discussion The effectiveness of input reconstruction reliable estimation stems from the fact that the network has a small number of hidden units and is only trained in a narrow range of situations. These constraints prevent the network from faithfully encoding arbitrary input patterns. Instead, the hidden units learn to encode features in the training images that are most important for the task. Baldi and Hornik [Baldi & Hornik, 1989] have shown that if an autoencoder
7 Input Reconstruction Reliability Estimation 285 network with a single layer of N linear hidden units is trained with back-propagation, the activation levels of the hidden units will represent the first N principal components of the training set. Since the units in the driving network are non-linear, this assertion does not strictly hold in this case. However, Cottrell and Munro [Cottrell & Munro, 1988] have found empirically that autoencoder networks with a sigmoidal activation function develop hidden units that span the principal subspace of the training images, with some noise on the first principal component due to network non-linearity. Because the principal components represent the dimensions along which the training examples varies most, it can be shown that using linear combinations of the principal components to represent the individual training patterns optimally preserves the information contained in the training set [Linsker, 1989]. However the compressed representation developed by a linear autoencoder network is only optimal for encoding images from the same distribution as the training set. When presented with images very different from those in the training set, the image reconstructed from the internal representation is not as accurate. The results presented in this paper demonstrate that this reconstruction error can be employed to estimate the likelihood and magnitude of error in MLPs trained for autonomous driving. However the input reconstruction technique presented here has a serious potential shortcoming, namely that it forces the network's hidden units to encode all input features, including potentially irrelevant ones. While this increased representation load on the hidden units has the potential to degrade network performance, this effect has not been observed in the tests conducted so far. In support of this finding, Gluck [Gluck, personal communications] has found that forcing a network to autoencode its input frequently improves its generalization. In [Gluck & Myers, 1992], Gluck and Myers use the representation developed by an autoencoder network as a model for simple types of learning in biological systems. The model suggests that the hippocampus acts as an autoencoder, developing internal representations that are then used to perform other tasks. But if interference from the autoencoder task proves to be a problem, one way to eliminate it would be to have separate groups of hidden units connected exclusively to one group of outputs or the other. Having a separate set of hidden units for the autoencoder task would ensure that the representation developed for the input reconstruction does not interfere with representation developed for the "normal" task. It remains to be seen if this decoupling of internal representations will adversely affect IRRE's ability to predict network errors. As a technique for multi-network integration, IRRE has several advantages over existing connectionist arbitration methods, such as Hampshire and Waibel's Meta-Pi architecture [Hampshire & Waibel, 1992] and the Adaptive Mixture of Experts Model of Jacobs et ai. [Jacobs et al., 1991]. It is a more modular approach, since each expert can be trained entirely in isolation and then later combined with other experts without any additional training by simply selecting the most reliable network for the current input. Since IRRE provides an absolute measure of a single network's reliability, and not just a measure of how appropriate a network is relative to others, IRRE can be also used to determine when none of the experts is capable of coping with the current situation. A potentially interesting extension to IRRE is the development of techniques for reasoning about the difference between the actual input and the reconstructed input. For instance, it should be possible to recognize when the vehicle has reached a fork in the road by the characteristic mistakes the network makes in reconstructing the input image. Another important component of future work in is to test the ability of IRRE to estimate network reliability in domains other than autonomous driving.
8 286 Pomerleau Acknowledgements I thank Dave Touretzky, Chuck Thorpe and the entire Unmanned Ground Vehicle Group at CMU for their support and suggestions. Principle support for this research has come from DARPA, under contracts "Perception for Outdoor Navigation" (contract number DACA76-89-C-OOI4, monitored by the US Army Topographic Engineering Center) and ''Unmanned Ground Vehicle System" (contract number DAAE07-90-C-R059, monitored by TACOM). References [Baldi & Hornik,1989] Baldi, P. and Hornik, K. (1989) Neural networks and principal component analysis: Learning from examples without local minima. Neural Networks, Vol. 2 pp [Cottrell & Munro, 1988] Cottrell, G.W. and Munro, P. (1988) Principal components analysis ofimages via back-propagation. Proc. Soc. of Photo-Opticallnstr. Eng., Cambridge MA. [Gluck, personal communications] Gluck, M.A. (1992) Personal Communications. Rutgers Univ., Newark NJ. [Gluck & Myers, 1992] Gluck, M.A. and Myers, C.E. (1992) Hippocampal function in representation and generalization: a computational theory. Proc.1992 Cogn. Sci. Soc. Con/. Hillsdale, NJ: Erlbaum Assoc. [Hampshire & Waibel, 1992] Hampshire J.B. and Waibel, A.H. (1989) The Meta-Pi network: building distributed knowledge representations for robust pattern recognition. IEEE Trans. on Pattern Analysis and Machine Intelligence. [Jacobs et al., 1991] Jacobs, R.A., Jordan, M.I. Nowlan, SJ. and Hinton, G.E. (1991) Adaptive mixtures of local experts. Neural Computation, 3 :1, Terrence Sejnowski (ed). [Linsker, 1989] Linsker, R. (1989) Designing a sensory processing system: What can be learned from principal component analysis? IBM Technical Report RC (#66896). [pomerleau, 1991] Pomerleau, D.A. (1991) Efficient Training of Artificial Neural Networks for Autonomous Navigation. Neural Computation 3:1, Terrence Sejnowski (ed). [pomerleau et al., 1991] Pomerleau, D.A., Gowdy, J., Thorpe, C.E. (1991) Combining artificial neural networks and symbolic processing for autonomous robot guidance. Engineering Applications of Artificial Intelligence, 4:4 pp
Rapidly Adapting Artificial Neural Networks for Autonomous Navigation
Rapidly Adapting Artificial Neural Networks for Autonomous Navigation Dean A. Pomerleau School of Computer Science Carnegie Mellon University Pittsburgh, PA 15213 Abstract The ALVINN (Autonomous Land Vehicle
More informationCombining Artificial Neural Networks and Symbolic Processing for Autonomous Robot Guidance
. ~ ~ Engng App/ic. ArliJ. Inrell. Vol. 4. No. 4, pp, 279-285, 1991 Printed in Grcat Bntain. All rights rcscrved OYS~-IY~~/YI $~.o()+o.oo Copyright 01991 Pcrgamon Prcss plc Contributed Paper Combining
More informationNetworks for Autonomous Navigation. Dean A. Pomerleau. Carnegie Mellon University. Pittsburgh, PA Abstract
Ecient Training of Articial Neural Networks for Autonomous Navigation Dean A. Pomerleau School of Computer Science Carnegie Mellon University Pittsburgh, PA 15213 Abstract The ALVINN (Autonomous Land Vehicle
More informationEfficient Training of Artificial Neural Networks for Autonomous Navigation
Communicated bv Dana Ballard Efficient Training of Artificial Neural Networks for Autonomous Navigation Dean A. Pomerleau School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213 USA
More informationEnhanced MLP Input-Output Mapping for Degraded Pattern Recognition
Enhanced MLP Input-Output Mapping for Degraded Pattern Recognition Shigueo Nomura and José Ricardo Gonçalves Manzan Faculty of Electrical Engineering, Federal University of Uberlândia, Uberlândia, MG,
More informationLearning New Articulator Trajectories for a Speech Production Model using Artificial Neural Networks
Learning New Articulator Trajectories for a Speech Production Model using Artificial Neural Networks C. S. Blackburn and S. J. Young Cambridge University Engineering Department (CUED), England email: csb@eng.cam.ac.uk
More informationFigure 1. Artificial Neural Network structure. B. Spiking Neural Networks Spiking Neural networks (SNNs) fall into the third generation of neural netw
Review Analysis of Pattern Recognition by Neural Network Soni Chaturvedi A.A.Khurshid Meftah Boudjelal Electronics & Comm Engg Electronics & Comm Engg Dept. of Computer Science P.I.E.T, Nagpur RCOEM, Nagpur
More informationStanford Center for AI Safety
Stanford Center for AI Safety Clark Barrett, David L. Dill, Mykel J. Kochenderfer, Dorsa Sadigh 1 Introduction Software-based systems play important roles in many areas of modern life, including manufacturing,
More informationINTELLIGENT UNMANNED GROUND VEHICLES Autonomous Navigation Research at Carnegie Mellon
INTELLIGENT UNMANNED GROUND VEHICLES Autonomous Navigation Research at Carnegie Mellon THE KLUWER INTERNATIONAL SERIES IN ENGINEERING AND COMPUTER SCIENCE ROBOTICS: VISION, MANIPULATION AND SENSORS Consulting
More informationTransactions on Information and Communications Technologies vol 1, 1993 WIT Press, ISSN
Combining multi-layer perceptrons with heuristics for reliable control chart pattern classification D.T. Pham & E. Oztemel Intelligent Systems Research Laboratory, School of Electrical, Electronic and
More informationAutonomous Systems. Robotics and. Reliability estimation for neural network based autonomous driving. Reprinted trom. Dean A.
Reprinted trom Robotics and Autonomous Systems Robotics and Autonomous Systems 12 (1994) 113-119 Reliability estimation for neural network based autonomous driving Dean A. Pomerleau School of Computer
More informationAn Autonomous Mobile Robot Architecture Using Belief Networks and Neural Networks
An Autonomous Mobile Robot Architecture Using Belief Networks and Neural Networks Mehran Sahami, John Lilly and Bryan Rollins Computer Science Department Stanford University Stanford, CA 94305 {sahami,lilly,rollins}@cs.stanford.edu
More informationA Neural Network Facial Expression Recognition System using Unsupervised Local Processing
A Neural Network Facial Expression Recognition System using Unsupervised Local Processing Leonardo Franco Alessandro Treves Cognitive Neuroscience Sector - SISSA 2-4 Via Beirut, Trieste, 34014 Italy lfranco@sissa.it,
More informationIntroduction to Machine Learning
Introduction to Machine Learning Deep Learning Barnabás Póczos Credits Many of the pictures, results, and other materials are taken from: Ruslan Salakhutdinov Joshua Bengio Geoffrey Hinton Yann LeCun 2
More informationAnalysis of Learning Paradigms and Prediction Accuracy using Artificial Neural Network Models
Analysis of Learning Paradigms and Prediction Accuracy using Artificial Neural Network Models Poornashankar 1 and V.P. Pawar 2 Abstract: The proposed work is related to prediction of tumor growth through
More informationNeural Networks and Antenna Arrays
Neural Networks and Antenna Arrays MAJA SAREVSKA 1, NIKOS MASTORAKIS 2 1 Istanbul Technical University, Istanbul, TURKEY 2 Hellenic Naval Academy, Athens, GREECE sarevska@itu.edu.tr mastor@wseas.org Abstract:
More informationAn Hybrid MLP-SVM Handwritten Digit Recognizer
An Hybrid MLP-SVM Handwritten Digit Recognizer A. Bellili ½ ¾ M. Gilloux ¾ P. Gallinari ½ ½ LIP6, Université Pierre et Marie Curie ¾ La Poste 4, Place Jussieu 10, rue de l Ile Mabon, BP 86334 75252 Paris
More informationThe Basic Kak Neural Network with Complex Inputs
The Basic Kak Neural Network with Complex Inputs Pritam Rajagopal The Kak family of neural networks [3-6,2] is able to learn patterns quickly, and this speed of learning can be a decisive advantage over
More informationGESTURE RECOGNITION FOR ROBOTIC CONTROL USING DEEP LEARNING
2017 NDIA GROUND VEHICLE SYSTEMS ENGINEERING AND TECHNOLOGY SYMPOSIUM AUTONOMOUS GROUND SYSTEMS (AGS) TECHNICAL SESSION AUGUST 8-10, 2017 - NOVI, MICHIGAN GESTURE RECOGNITION FOR ROBOTIC CONTROL USING
More informationBehaviour-Based Control. IAR Lecture 5 Barbara Webb
Behaviour-Based Control IAR Lecture 5 Barbara Webb Traditional sense-plan-act approach suggests a vertical (serial) task decomposition Sensors Actuators perception modelling planning task execution motor
More informationFusing Radar and Vision for Detecting, Classifying and Avoiding Roadway Obstacles
ntelligent Vehicles, Tokyo, Japan, September 18-20, 199t Fusing Radar and Vision for Detecting, Classifying and Avoiding Roadway Obstacles Dirk Langer and Todd Jochem The Robotics nstitute, Carnegie Mellon
More informationApplication of Multi Layer Perceptron (MLP) for Shower Size Prediction
Chapter 3 Application of Multi Layer Perceptron (MLP) for Shower Size Prediction 3.1 Basic considerations of the ANN Artificial Neural Network (ANN)s are non- parametric prediction tools that can be used
More informationTarget Classification in Forward Scattering Radar in Noisy Environment
Target Classification in Forward Scattering Radar in Noisy Environment Mohamed Khala Alla H.M, Mohamed Kanona and Ashraf Gasim Elsid School of telecommunication and space technology, Future university
More informationMAGNT Research Report (ISSN ) Vol.6(1). PP , Controlling Cost and Time of Construction Projects Using Neural Network
Controlling Cost and Time of Construction Projects Using Neural Network Li Ping Lo Faculty of Computer Science and Engineering Beijing University China Abstract In order to achieve optimized management,
More informationUsing RASTA in task independent TANDEM feature extraction
R E S E A R C H R E P O R T I D I A P Using RASTA in task independent TANDEM feature extraction Guillermo Aradilla a John Dines a Sunil Sivadas a b IDIAP RR 04-22 April 2004 D a l l e M o l l e I n s t
More informationSupplementary information accompanying the manuscript Biologically Inspired Modular Neural Control for a Leg-Wheel Hybrid Robot
Supplementary information accompanying the manuscript Biologically Inspired Modular Neural Control for a Leg-Wheel Hybrid Robot Poramate Manoonpong a,, Florentin Wörgötter a, Pudit Laksanacharoen b a)
More informationAN ANN BASED FAULT DETECTION ON ALTERNATOR
AN ANN BASED FAULT DETECTION ON ALTERNATOR Suraj J. Dhon 1, Sarang V. Bhonde 2 1 (Electrical engineering, Amravati University, India) 2 (Electrical engineering, Amravati University, India) ABSTRACT: Synchronous
More informationNOTICE WARNING CONCERNING COPYRIGHT RESTRICTIONS: The copyright law of the United States (title 17, U.S. Code) governs the making of photocopies or
NOTICE WARNING CONCERNING COPYRIGHT RESTRICTIONS: The copyright law of the United States (title 17, U.S. Code) governs the making of photocopies or other reproductions of copyrighted material. Any copying
More informationA Comparison of Particle Swarm Optimization and Gradient Descent in Training Wavelet Neural Network to Predict DGPS Corrections
Proceedings of the World Congress on Engineering and Computer Science 00 Vol I WCECS 00, October 0-, 00, San Francisco, USA A Comparison of Particle Swarm Optimization and Gradient Descent in Training
More informationSurveillance and Calibration Verification Using Autoassociative Neural Networks
Surveillance and Calibration Verification Using Autoassociative Neural Networks Darryl J. Wrest, J. Wesley Hines, and Robert E. Uhrig* Department of Nuclear Engineering, University of Tennessee, Knoxville,
More informationA comparative study of different feature sets for recognition of handwritten Arabic numerals using a Multi Layer Perceptron
Proc. National Conference on Recent Trends in Intelligent Computing (2006) 86-92 A comparative study of different feature sets for recognition of handwritten Arabic numerals using a Multi Layer Perceptron
More informationApplication of Artificial Neural Networks in Autonomous Mission Planning for Planetary Rovers
Application of Artificial Neural Networks in Autonomous Mission Planning for Planetary Rovers 1 Institute of Deep Space Exploration Technology, School of Aerospace Engineering, Beijing Institute of Technology,
More informationNeuro-Fuzzy and Soft Computing: Fuzzy Sets. Chapter 1 of Neuro-Fuzzy and Soft Computing by Jang, Sun and Mizutani
Chapter 1 of Neuro-Fuzzy and Soft Computing by Jang, Sun and Mizutani Outline Introduction Soft Computing (SC) vs. Conventional Artificial Intelligence (AI) Neuro-Fuzzy (NF) and SC Characteristics 2 Introduction
More informationPose Invariant Face Recognition
Pose Invariant Face Recognition Fu Jie Huang Zhihua Zhou Hong-Jiang Zhang Tsuhan Chen Electrical and Computer Engineering Department Carnegie Mellon University jhuangfu@cmu.edu State Key Lab for Novel
More informationHighly-Accurate Real-Time GPS Carrier Phase Disciplined Oscillator
Highly-Accurate Real-Time GPS Carrier Phase Disciplined Oscillator C.-L. Cheng, F.-R. Chang, L.-S. Wang, K.-Y. Tu Dept. of Electrical Engineering, National Taiwan University. Inst. of Applied Mechanics,
More informationAGENT PLATFORM FOR ROBOT CONTROL IN REAL-TIME DYNAMIC ENVIRONMENTS. Nuno Sousa Eugénio Oliveira
AGENT PLATFORM FOR ROBOT CONTROL IN REAL-TIME DYNAMIC ENVIRONMENTS Nuno Sousa Eugénio Oliveira Faculdade de Egenharia da Universidade do Porto, Portugal Abstract: This paper describes a platform that enables
More informationA Foveated Visual Tracking Chip
TP 2.1: A Foveated Visual Tracking Chip Ralph Etienne-Cummings¹, ², Jan Van der Spiegel¹, ³, Paul Mueller¹, Mao-zhu Zhang¹ ¹Corticon Inc., Philadelphia, PA ²Department of Electrical Engineering, Southern
More informationArtificial Neural Network based Mobile Robot Navigation
Artificial Neural Network based Mobile Robot Navigation István Engedy Budapest University of Technology and Economics, Department of Measurement and Information Systems, Magyar tudósok körútja 2. H-1117,
More informationImage Enhancement in Spatial Domain
Image Enhancement in Spatial Domain 2 Image enhancement is a process, rather a preprocessing step, through which an original image is made suitable for a specific application. The application scenarios
More informationENVIRONMENTALLY ADAPTIVE SONAR CONTROL IN A TACTICAL SETTING
ENVIRONMENTALLY ADAPTIVE SONAR CONTROL IN A TACTICAL SETTING WARREN L. J. FOX, MEGAN U. HAZEN, AND CHRIS J. EGGEN University of Washington, Applied Physics Laboratory, 13 NE 4th St., Seattle, WA 98, USA
More informationCS295-1 Final Project : AIBO
CS295-1 Final Project : AIBO Mert Akdere, Ethan F. Leland December 20, 2005 Abstract This document is the final report for our CS295-1 Sensor Data Management Course Final Project: Project AIBO. The main
More informationPolicy Forum. Science 26 January 2001: Vol no. 5504, pp DOI: /science Prev Table of Contents Next
Science 26 January 2001: Vol. 291. no. 5504, pp. 599-600 DOI: 10.1126/science.291.5504.599 Prev Table of Contents Next Policy Forum ARTIFICIAL INTELLIGENCE: Autonomous Mental Development by Robots and
More informationAN IMPROVED NEURAL NETWORK-BASED DECODER SCHEME FOR SYSTEMATIC CONVOLUTIONAL CODE. A Thesis by. Andrew J. Zerngast
AN IMPROVED NEURAL NETWORK-BASED DECODER SCHEME FOR SYSTEMATIC CONVOLUTIONAL CODE A Thesis by Andrew J. Zerngast Bachelor of Science, Wichita State University, 2008 Submitted to the Department of Electrical
More informationNeural Network Driving with dierent Sensor Types in a Virtual Environment
Neural Network Driving with dierent Sensor Types in a Virtual Environment Postgraduate Project Department of Computer Science University of Auckland New Zealand Benjamin Seidler supervised by Dr Burkhard
More informationThe next level of intelligence: Artificial Intelligence. Innovation Day USA 2017 Princeton, March 27, 2017 Michael May, Siemens Corporate Technology
The next level of intelligence: Artificial Intelligence Innovation Day USA 2017 Princeton, March 27, 2017, Siemens Corporate Technology siemens.com/innovationusa Notes and forward-looking statements This
More informationChapter 1: Introduction to Neuro-Fuzzy (NF) and Soft Computing (SC)
Chapter 1: Introduction to Neuro-Fuzzy (NF) and Soft Computing (SC) Introduction (1.1) SC Constituants and Conventional Artificial Intelligence (AI) (1.2) NF and SC Characteristics (1.3) Jyh-Shing Roger
More informationHigh-speed Noise Cancellation with Microphone Array
Noise Cancellation a Posteriori Probability, Maximum Criteria Independent Component Analysis High-speed Noise Cancellation with Microphone Array We propose the use of a microphone array based on independent
More informationAdvanced Robotics Introduction
Advanced Robotics Introduction Institute for Software Technology 1 Motivation Agenda Some Definitions and Thought about Autonomous Robots History Challenges Application Examples 2 http://youtu.be/rvnvnhim9kg
More informationIncorporating a Connectionist Vision Module into a Fuzzy, Behavior-Based Robot Controller
From:MAICS-97 Proceedings. Copyright 1997, AAAI (www.aaai.org). All rights reserved. Incorporating a Connectionist Vision Module into a Fuzzy, Behavior-Based Robot Controller Douglas S. Blank and J. Oliver
More informationDeveloping Frogger Player Intelligence Using NEAT and a Score Driven Fitness Function
Developing Frogger Player Intelligence Using NEAT and a Score Driven Fitness Function Davis Ancona and Jake Weiner Abstract In this report, we examine the plausibility of implementing a NEAT-based solution
More informationApplication of Generalised Regression Neural Networks in Lossless Data Compression
Application of Generalised Regression Neural Networks in Lossless Data Compression R. LOGESWARAN Centre for Multimedia Communications, Faculty of Engineering, Multimedia University, 63100 Cyberjaya MALAYSIA
More informationTurboDrive. With the recent introduction of the Linea GigE line scan cameras, Teledyne DALSA is once again pushing innovation to new heights.
With the recent introduction of the Linea GigE line scan cameras, Teledyne DALSA is once again pushing innovation to new heights. The Linea GigE is the first Teledyne DALSA camera to offer. This technology
More informationFuzzy Logic Based Robot Navigation In Uncertain Environments By Multisensor Integration
Proceedings of the 1994 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MF1 94) Las Vega, NV Oct. 2-5, 1994 Fuzzy Logic Based Robot Navigation In Uncertain
More informationLane Detection in Automotive
Lane Detection in Automotive Contents Introduction... 2 Image Processing... 2 Reading an image... 3 RGB to Gray... 3 Mean and Gaussian filtering... 5 Defining our Region of Interest... 6 BirdsEyeView Transformation...
More informationA Robust Neural Robot Navigation Using a Combination of Deliberative and Reactive Control Architectures
A Robust Neural Robot Navigation Using a Combination of Deliberative and Reactive Control Architectures D.M. Rojas Castro, A. Revel and M. Ménard * Laboratory of Informatics, Image and Interaction (L3I)
More informationAn Autonomous Vehicle Navigation System using Panoramic Machine Vision Techniques
An Autonomous Vehicle Navigation System using Panoramic Machine Vision Techniques Kevin Rushant, Department of Computer Science, University of Sheffield, GB. email: krusha@dcs.shef.ac.uk Libor Spacek,
More informationCHAPTER 8: EXTENDED TETRACHORD CLASSIFICATION
CHAPTER 8: EXTENDED TETRACHORD CLASSIFICATION Chapter 7 introduced the notion of strange circles: using various circles of musical intervals as equivalence classes to which input pitch-classes are assigned.
More informationIntroduction to Machine Learning
Introduction to Machine Learning Perceptron Barnabás Póczos Contents History of Artificial Neural Networks Definitions: Perceptron, Multi-Layer Perceptron Perceptron algorithm 2 Short History of Artificial
More informationSupervised Versus Unsupervised Binary-Learning by Feedforward Neural Networks
Machine Learning, 42, 97 122, 2001 c 2001 Kluwer Academic Publishers. Manufactured in The Netherlands. Supervised Versus Unsupervised Binary-Learning by Feedforward Neural Networks NATHALIE JAPKOWICZ nat@site.uottawa.ca
More informationEvolved Neurodynamics for Robot Control
Evolved Neurodynamics for Robot Control Frank Pasemann, Martin Hülse, Keyan Zahedi Fraunhofer Institute for Autonomous Intelligent Systems (AiS) Schloss Birlinghoven, D-53754 Sankt Augustin, Germany Abstract
More informationLearning and Using Models of Kicking Motions for Legged Robots
Learning and Using Models of Kicking Motions for Legged Robots Sonia Chernova and Manuela Veloso Computer Science Department Carnegie Mellon University Pittsburgh, PA 15213 {soniac, mmv}@cs.cmu.edu Abstract
More informationMINE 432 Industrial Automation and Robotics
MINE 432 Industrial Automation and Robotics Part 3, Lecture 5 Overview of Artificial Neural Networks A. Farzanegan (Visiting Associate Professor) Fall 2014 Norman B. Keevil Institute of Mining Engineering
More informationArrangement of Robot s sonar range sensors
MOBILE ROBOT SIMULATION BY MEANS OF ACQUIRED NEURAL NETWORK MODELS Ten-min Lee, Ulrich Nehmzow and Roger Hubbold Department of Computer Science, University of Manchester Oxford Road, Manchester M 9PL,
More informationMULTIPLE CLASSIFIERS FOR ELECTRONIC NOSE DATA
MULTIPLE CLASSIFIERS FOR ELECTRONIC NOSE DATA M. Pardo, G. Sberveglieri INFM and University of Brescia Gas Sensor Lab, Dept. of Chemistry and Physics for Materials Via Valotti 9-25133 Brescia Italy D.
More informationApplication of Artificial Neural Networks System for Synthesis of Phased Cylindrical Arc Antenna Arrays
International Journal of Communication Engineering and Technology. ISSN 2277-3150 Volume 4, Number 1 (2014), pp. 7-15 Research India Publications http://www.ripublication.com Application of Artificial
More informationOptimal Rhode Island Hold em Poker
Optimal Rhode Island Hold em Poker Andrew Gilpin and Tuomas Sandholm Computer Science Department Carnegie Mellon University Pittsburgh, PA 15213 {gilpin,sandholm}@cs.cmu.edu Abstract Rhode Island Hold
More informationImproving the Safety and Efficiency of Roadway Maintenance Phase II: Developing a Vision Guidance System for the Robotic Roadway Message Painter
Improving the Safety and Efficiency of Roadway Maintenance Phase II: Developing a Vision Guidance System for the Robotic Roadway Message Painter Final Report Prepared by: Ryan G. Rosandich Department of
More informationUser interface for remote control robot
User interface for remote control robot Gi-Oh Kim*, and Jae-Wook Jeon ** * Department of Electronic and Electric Engineering, SungKyunKwan University, Suwon, Korea (Tel : +8--0-737; E-mail: gurugio@ece.skku.ac.kr)
More informationHuman-robot relation. Human-robot relation
Town Robot { Toward social interaction technologies of robot systems { Hiroshi ISHIGURO and Katsumi KIMOTO Department of Information Science Kyoto University Sakyo-ku, Kyoto 606-01, JAPAN Email: ishiguro@kuis.kyoto-u.ac.jp
More informationNonlinear System Identification Using Recurrent Networks
Syracuse University SURFACE Electrical Engineering and Computer Science Technical Reports College of Engineering and Computer Science 7-1991 Nonlinear System Identification Using Recurrent Networks Hyungkeun
More informationAVCS Research at Carnegie Mellon University
AVCS Research at Carnegie Mellon University Dean Pomerleau, Charles Thorpe, Dirk Langer, Julio K. Rosenblatt and Rahul Sukthankar Robotics Institute Carnegie Mellon University Pittsburgh PA USA Abstract:
More informationMel Spectrum Analysis of Speech Recognition using Single Microphone
International Journal of Engineering Research in Electronics and Communication Mel Spectrum Analysis of Speech Recognition using Single Microphone [1] Lakshmi S.A, [2] Cholavendan M [1] PG Scholar, Sree
More informationCreating a 3D environment map from 2D camera images in robotics
Creating a 3D environment map from 2D camera images in robotics J.P. Niemantsverdriet jelle@niemantsverdriet.nl 4th June 2003 Timorstraat 6A 9715 LE Groningen student number: 0919462 internal advisor:
More informationIntelligent Technology for More Advanced Autonomous Driving
FEATURED ARTICLES Autonomous Driving Technology for Connected Cars Intelligent Technology for More Advanced Autonomous Driving Autonomous driving is recognized as an important technology for dealing with
More informationIntroduction to Vision & Robotics
Introduction to Vision & Robotics Vittorio Ferrari, 650-2697,IF 1.27 vferrari@staffmail.inf.ed.ac.uk Michael Herrmann, 651-7177, IF1.42 mherrman@inf.ed.ac.uk Lectures: Handouts will be on the web (but
More informationEvaluation of Waveform Structure Features on Time Domain Target Recognition under Cross Polarization
Journal of Physics: Conference Series PAPER OPEN ACCESS Evaluation of Waveform Structure Features on Time Domain Target Recognition under Cross Polarization To cite this article: M A Selver et al 2016
More information8.2 IMAGE PROCESSING VERSUS IMAGE ANALYSIS Image processing: The collection of routines and
8.1 INTRODUCTION In this chapter, we will study and discuss some fundamental techniques for image processing and image analysis, with a few examples of routines developed for certain purposes. 8.2 IMAGE
More informationMaze Solving Algorithms for Micro Mouse
Maze Solving Algorithms for Micro Mouse Surojit Guha Sonender Kumar surojitguha1989@gmail.com sonenderkumar@gmail.com Abstract The problem of micro-mouse is 30 years old but its importance in the field
More informationNeural Networks for Real-time Pathfinding in Computer Games
Neural Networks for Real-time Pathfinding in Computer Games Ross Graham 1, Hugh McCabe 1 & Stephen Sheridan 1 1 School of Informatics and Engineering, Institute of Technology at Blanchardstown, Dublin
More informationAdvanced Robotics Introduction
Advanced Robotics Introduction Institute for Software Technology 1 Agenda Motivation Some Definitions and Thought about Autonomous Robots History Challenges Application Examples 2 Bridge the Gap Mobile
More informationTadeusz Stepinski and Bengt Vagnhammar, Uppsala University, Signals and Systems, Box 528, SE Uppsala, Sweden
AUTOMATIC DETECTING DISBONDS IN LAYERED STRUCTURES USING ULTRASONIC PULSE-ECHO INSPECTION Tadeusz Stepinski and Bengt Vagnhammar, Uppsala University, Signals and Systems, Box 58, SE-751 Uppsala, Sweden
More informationHISTOGRAM BASED AUTOMATIC IMAGE SEGMENTATION USING WAVELETS FOR IMAGE ANALYSIS
HISTOGRAM BASED AUTOMATIC IMAGE SEGMENTATION USING WAVELETS FOR IMAGE ANALYSIS Samireddy Prasanna 1, N Ganesh 2 1 PG Student, 2 HOD, Dept of E.C.E, TPIST, Komatipalli, Bobbili, Andhra Pradesh, (India)
More informationA Bottom-Up Approach to on-chip Signal Integrity
A Bottom-Up Approach to on-chip Signal Integrity Andrea Acquaviva, and Alessandro Bogliolo Information Science and Technology Institute (STI) University of Urbino 6029 Urbino, Italy acquaviva@sti.uniurb.it
More informationMAGNETORESISTIVE random access memory
132 IEEE TRANSACTIONS ON MAGNETICS, VOL. 41, NO. 1, JANUARY 2005 A 4-Mb Toggle MRAM Based on a Novel Bit and Switching Method B. N. Engel, J. Åkerman, B. Butcher, R. W. Dave, M. DeHerrera, M. Durlam, G.
More informationDigital image processing vs. computer vision Higher-level anchoring
Digital image processing vs. computer vision Higher-level anchoring Václav Hlaváč Czech Technical University in Prague Faculty of Electrical Engineering, Department of Cybernetics Center for Machine Perception
More informationChapter - 7. Adaptive Channel Equalization
Chapter - 7 Adaptive Channel Equalization Chapter - 7 Adaptive Channel Equalization 7.1 Introduction The transmission o f digital information over a communication channel causes Inter Symbol Interference
More informationUsing Neural Network and Monte-Carlo Tree Search to Play the Game TEN
Using Neural Network and Monte-Carlo Tree Search to Play the Game TEN Weijie Chen Fall 2017 Weijie Chen Page 1 of 7 1. INTRODUCTION Game TEN The traditional game Tic-Tac-Toe enjoys people s favor. Moreover,
More informationThe Future of AI A Robotics Perspective
The Future of AI A Robotics Perspective Wolfram Burgard Autonomous Intelligent Systems Department of Computer Science University of Freiburg Germany The Future of AI My Robotics Perspective Wolfram Burgard
More informationLong Range Acoustic Classification
Approved for public release; distribution is unlimited. Long Range Acoustic Classification Authors: Ned B. Thammakhoune, Stephen W. Lang Sanders a Lockheed Martin Company P. O. Box 868 Nashua, New Hampshire
More informationKeywords: Multi-robot adversarial environments, real-time autonomous robots
ROBOT SOCCER: A MULTI-ROBOT CHALLENGE EXTENDED ABSTRACT Manuela M. Veloso School of Computer Science Carnegie Mellon University Pittsburgh, PA 15213, USA veloso@cs.cmu.edu Abstract Robot soccer opened
More informationIntegrated Vision and Sound Localization
Integrated Vision and Sound Localization Parham Aarabi Safwat Zaky Department of Electrical and Computer Engineering University of Toronto 10 Kings College Road, Toronto, Ontario, Canada, M5S 3G4 parham@stanford.edu
More informationApproximation a One-Dimensional Functions by Using Multilayer Perceptron and Radial Basis Function Networks
Approximation a One-Dimensional Functions by Using Multilayer Perceptron and Radial Basis Function Networks Huda Dheyauldeen Najeeb Department of public relations College of Media, University of Al Iraqia,
More informationExploration and Model Building in Mobile Robot Domains
In: Proceedings of the IEEE International Conference on Neural Networks San Francisco, CA, March 28-April 1, 1993 Exploration and Model Building in Mobile Robot Domains Sebastian B. Thrun School of Computer
More informationMULTI-LAYERED HYBRID ARCHITECTURE TO SOLVE COMPLEX TASKS OF AN AUTONOMOUS MOBILE ROBOT
MULTI-LAYERED HYBRID ARCHITECTURE TO SOLVE COMPLEX TASKS OF AN AUTONOMOUS MOBILE ROBOT F. TIECHE, C. FACCHINETTI and H. HUGLI Institute of Microtechnology, University of Neuchâtel, Rue de Tivoli 28, CH-2003
More informationCLASSIFICATION OF CLOSED AND OPEN-SHELL (TURKISH) PISTACHIO NUTS USING DOUBLE TREE UN-DECIMATED WAVELET TRANSFORM
CLASSIFICATION OF CLOSED AND OPEN-SHELL (TURKISH) PISTACHIO NUTS USING DOUBLE TREE UN-DECIMATED WAVELET TRANSFORM Nuri F. Ince 1, Fikri Goksu 1, Ahmed H. Tewfik 1, Ibrahim Onaran 2, A. Enis Cetin 2, Tom
More informationNCCT IEEE PROJECTS ADVANCED ROBOTICS SOLUTIONS. Latest Projects, in various Domains. Promise for the Best Projects
NCCT Promise for the Best Projects IEEE PROJECTS in various Domains Latest Projects, 2009-2010 ADVANCED ROBOTICS SOLUTIONS EMBEDDED SYSTEM PROJECTS Microcontrollers VLSI DSP Matlab Robotics ADVANCED ROBOTICS
More informationImplicit Fitness Functions for Evolving a Drawing Robot
Implicit Fitness Functions for Evolving a Drawing Robot Jon Bird, Phil Husbands, Martin Perris, Bill Bigge and Paul Brown Centre for Computational Neuroscience and Robotics University of Sussex, Brighton,
More informationResearch Statement MAXIM LIKHACHEV
Research Statement MAXIM LIKHACHEV My long-term research goal is to develop a methodology for robust real-time decision-making in autonomous systems. To achieve this goal, my students and I research novel
More informationArtificial Beacons with RGB-D Environment Mapping for Indoor Mobile Robot Localization
Sensors and Materials, Vol. 28, No. 6 (2016) 695 705 MYU Tokyo 695 S & M 1227 Artificial Beacons with RGB-D Environment Mapping for Indoor Mobile Robot Localization Chun-Chi Lai and Kuo-Lan Su * Department
More informationEvolving High-Dimensional, Adaptive Camera-Based Speed Sensors
In: M.H. Hamza (ed.), Proceedings of the 21st IASTED Conference on Applied Informatics, pp. 1278-128. Held February, 1-1, 2, Insbruck, Austria Evolving High-Dimensional, Adaptive Camera-Based Speed Sensors
More information